JP2016122325A - Diagnosis support system image processing method and program of the diagnosis support system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a diagnosis support system capable of supporting and facilitating a diagnosis made by a medical doctor while improving the diagnosis accuracy.SOLUTION: A processing section 101 has an extraction means 101b which includes at least one of: segmentation means for segmenting a pick-up image based on color quality; first extraction means 101b-1 that extracts a part candidate based on brightness components; and second extraction means 101b-2 that extracts a part candidate based on color space constituted of brightness element and color information element. The processing section selectively applies the first extraction means 101b-1 and the second extraction means 101b-2 according to the sort by a sorting means.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a diagnosis support apparatus, an image processing method in the diagnosis support apparatus, and a program thereof.

  Visual inspection is always performed as a diagnosis of skin lesions, and a lot of information can be obtained. However, even with the naked eye or loupe alone, it is difficult to distinguish between moles and spots, and it is also difficult to distinguish benign and malignant tumors. Therefore, dermoscopy diagnosis is performed in which lesions are imaged using a camera with a dermoscope.

  A dermoscope is a non-invasive diagnostic instrument that brightly illuminates a lesion with an LED lamp, a halogen lamp, or the like, and makes an echo gel or a polarizing filter free from reflected light to observe the magnified image about 10 times. The observation method using this instrument is called dermoscopy. The dermoscopy diagnosis is described in detail in the Internet URL (http://www.twmu.ac.jp/DNH/department/dermatology/dermoscopy.html) <browse September 1, 2014>. According to the dermoscopy diagnosis, the pigment distribution from the epidermis to the superficial dermis can be seen well by eliminating the irregular reflection due to the stratum corneum.

  For example, Patent Document 1 discloses a technique of a remote diagnosis system for a pigmentation site that performs diagnosis using skin tone images taken with the above-described dermoscope using values such as color tone, texture, asymmetry, and circularity. ing. It uses a camera phone with a dermoscope to photograph skin with benign pigmented nevus, which is a concern for melanoma, through the dermoscope. Then, the mobile phone is connected to the Internet using the network connection function of the mobile phone, and the photographed skin image is transmitted to the remote diagnosis support device to request diagnosis. The remote diagnosis support device that has received the skin image uses the melanoma diagnosis program to diagnose whether or not the skin image is a melanoma or, if it is a melanoma, the stage of the melanoma. The result is returned to the doctor.

JP-A-2005-192944

  By the way, an image photographed with a dermoscope is stored in a predetermined image format in accordance with the specifications of the dermoscope. Then, the stored dermoscope image is read out and displayed on the display device, or the dermoscope image is processed. However, depending on the storage format in which lossy compression such as JPEG is performed, detailed color information is lost. Sometimes. When browsing, there is no relatively great influence, but when diagnosing a lesion based on a minute change in color information, the lost color information may greatly affect the essence of the processing result.

  The present invention has been made in order to solve the above-described problems, and facilitates diagnosis by a doctor and improves diagnostic accuracy by highlighting and displaying a diagnosis target portion on an input image by an optimum method. It is an object of the present invention to provide a diagnosis support apparatus, an image processing method in the diagnosis support apparatus, and a program thereof.

In order to solve the above-described problem, a diagnosis support apparatus according to an aspect of the present invention is a diagnosis support apparatus for diagnosing a lesion using a captured image of an affected area, and an image storage unit that stores the captured image; A processing unit that processes the captured image stored in the image storage unit, wherein the processing unit separates the captured image into a luminance component and a color information component; Extracting means for extracting, a classifying means for classifying the captured image based on color quality, a first extracting means for extracting a region candidate by the luminance component, and a likelihood of a region as the luminance component. And at least one of second extraction means for extracting by a color space composed of the color information component, and selecting the first extraction means and the second extraction means according to the classification of the classification means To apply The features.
Other aspects of the present invention will become apparent from the description of this specification and the accompanying drawings.

  According to the present invention, a diagnosis support apparatus and an image in the diagnosis support apparatus that facilitate diagnosis of a doctor and improve diagnosis accuracy by highlighting and displaying a diagnosis target portion by a method optimal for an input image. A processing method and its program can be provided.

It is a block diagram which shows the structure of the diagnosis assistance apparatus which concerns on embodiment of this invention. It is a flowchart which shows the basic processing operation | movement of the diagnosis assistance apparatus which concerns on embodiment of this invention. 3 is a flowchart illustrating a blood vessel extraction processing operation of the captured image of FIG. 2. 4 is a flowchart showing a processing operation for setting a type of blood vessel extraction E processing according to the color quality of FIG. 3. It is a flowchart which shows the processing operation of the blood vessel extraction E processing type I of FIG. 4 is a flowchart showing an example of a processing operation for extracting the likelihood of blood vessels in FIG. 3 as likelihood A. FIG. 10 is a flowchart showing another example of the processing operation for extracting the likelihood of blood vessels in FIG. It is a flowchart which shows the processing operation of the blood vessel extraction E processing type II of FIG. It is a flowchart which shows the processing operation which performs blood-vessel extraction E from the likelihood which shows the blood-vessel likeness of FIG. It is a figure which shows an example of the display screen structure of the diagnosis assistance apparatus which concerns on embodiment of this invention.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention (hereinafter referred to as “embodiments”) will be described in detail with reference to the accompanying drawings. Note that the same number is assigned to the same element throughout the description of the embodiment.

(Configuration of the embodiment)
FIG. 1 is a block diagram illustrating a configuration of a diagnosis support apparatus 100 according to the present embodiment. As shown in FIG. 1, an imaging apparatus 110 with a dermoscope is connected to the diagnosis support apparatus 100 according to the present embodiment. The imaging device 110 with a dermoscope performs imaging in accordance with an instruction from the diagnosis support apparatus 100 (processing unit 101), stores the captured image (dermoscopy image) in the image storage unit 102, and displays the image on the display device 120. In addition, the captured image is emphasized by the processing unit 101 and stored in the image storage unit 102 and displayed on the display device 120. The input device 130 performs a dermoscope image capturing start instruction, a part selection operation in a dermoscopy image, which will be described later, and the like.

  The display device 120 is constituted by an LCD (Liquid Crystal Display) monitor, for example, and the input device 130 is constituted by a mouse or the like.

  The processing unit 101 processes a captured image stored in the image storage unit 102, and includes a separation unit 101a, an extraction unit 101b, and a generation unit 101c as shown in FIG.

  The separating unit 101a functions as a unit that separates a captured image into a luminance component and a color information component.

  The extraction unit 101b functions as a unit for extracting a part to be diagnosed, and classifies the captured image by the classifying unit based on the color quality, and also extracts a first extraction unit that extracts a part candidate based on the luminance component, and the part-likeness. It includes at least one second extraction unit that extracts a color space composed of a luminance component and a color information component, and the first extraction unit 101b-1 and the second extraction unit 101b-2 according to the section. Is applied selectively.

  Here, the classification is performed as follows. That is, the classification means sets the first classification when the captured image is an uncompressed image or an image from which color information is not thinned out, and sets the second classification when the captured image is thinned out from the compressed image or color information. Although details will be described later in the operation of the embodiment, the first section corresponds to when the color quality of the captured image is high, and the second section corresponds to when the color quality of the captured image is low.

  Depending on the color quality (low quality), the extraction unit 101b uses the luminance component to extract the region candidate by the first morphological process, and the second extraction unit 101b-2 A part-likeness may be extracted using a color space, and a part extraction image may be generated by integrating the part candidate and the part-likeness extracted by the extraction unit 101b.

  Depending on the color quality (high quality), the extracting unit 101b may extract the part-likeness using the color space by the second extracting unit 101b-2, and the second morphology using the part-likeness extracted by the extracting unit 101b. A site extraction image may be generated by processing.

  Here, the first morphological process is a closing process that repeats an expansion process and a contraction process in this order for the extracted luminance component, a smoothing filter process for the luminance component that has been subjected to the closing process, and a smoothing filter process. Subtraction processing for subtracting the luminance component of the photographed image from the luminance component. Also, as the second morphological process, an opening process for repeating the contraction process and the expansion process in this order for the extracted part-likeness, a smoothing filter process for the part-likeness subjected to the opening process, and a part subjected to the smoothing filter process The process of subtracting the likelihood from the extracted part likelihood is included.

  The separation means 101a and the extraction means 101b (the first extraction means 101b-1 and the second extraction means 101b-2) described above sequentially read and execute the program according to the present embodiment that the processing unit 101 has. As a result, the functions described above are realized.

(Operation of the embodiment)
Hereinafter, the operation of the diagnosis support apparatus 100 according to the present embodiment shown in FIG. 1 will be described in detail with reference to FIG.

  FIG. 2 shows a flow of basic processing operations of the diagnosis support apparatus 100 according to the present embodiment. According to FIG. 2, the processing unit 101 first obtains a photographed image of an affected part, such as a skin lesion site, photographed by the photographing apparatus 110 with a dermoscope (step S11). Then, the captured image acquired in a format according to the specification of the imaging device 110 with a dermoscope is stored in a predetermined area of the image storage unit 102 and displayed on the display device 120 (step S12).

  It is assumed that the photographed image is stored in a JPEG (Joint Photographic Experts Group) format used in many digital cameras. The JPEG format performs irreversible compression using the characteristics of human eyes, and can specify the resolution of color information (color resolution) when compression is performed. Many digital cameras use a sampling factor of 4: 2: 2. This means that the color information is thinned out in half with respect to the luminance information. A sampling factor of 4: 4: 4 can also be designated, which means that the resolution of luminance information (luminance resolution) and the resolution of color information (color resolution) are the same.

  Here, the color quality will be described. For example, an image with no information loss such as RAW data (RAW image format) or BMP data (Bit MaP image format) has high color quality, and an image with loss of information using irreversible compression like JPEG data. Is considered low color quality.

  As described above, JPEG data is obtained by thinning out the color resolution and partially cutting off the color frequency component. Therefore, color quality does not depend on the number of pixels or the performance of color processing, and does not mean that an image taken with 20 million pixels has a higher color quality than an image taken with 1 million pixels. RAW data is considered to have higher color quality than JPEG data in an image shot with 1 million pixels. Therefore, color quality is a concept different from image quality and color resolution. Naturally, RAW data has higher color quality and better image quality than JPEG data if it is an image taken with a camera of similar specifications.

  However, even with JPEG data, if the sampling factor is 4: 4: 4, the color resolution is not thinned out, so the color quality is considered high. In the following description of the embodiment, JPEG data having a sampling factor of 4: 4: 4 is described as an example. Therefore, it is assumed that the higher the color resolution of the image, the higher the color quality. On the other hand, JPEG data with a sampling factor of 4: 2: 2 is assumed to have low color quality.

  The diagnosis support apparatus 100 according to the present embodiment sets a type indicating the blood vessel extraction method according to the color resolution of the image (step S13), and performs blood vessel extraction processing according to the set type. That is, when the type I is set (step S14 “type I”), the processing unit 101 executes the blood vessel extraction process by the blood vessel extraction method I (step S15), and when the type II is set (step S14 “ The blood vessel extraction E processing by the type II ") blood vessel extraction method II is executed (step S16). Then, the processing result is displayed side by side on the display device 120 together with the captured image displayed earlier, and the diagnosis is left to the doctor (step S17).

  FIG. 10 shows an example of a display screen image displayed on the display device 120. A captured image display area 121 is allocated to the left and an emphasized image display area 122 is allocated to the right, and a captured image and a blood vessel emphasized image are displayed in each area. When the doctor uses the input device 130 and clicks the “start imaging” button 123 assigned to the lower right of the screen of the display device 120, imaging of the affected area is started by the imaging device 110 with a dermoscope. Then, by the blood vessel extraction processing by the processing unit 101, the captured image is displayed in the captured image display area 121 of the display device 120, and the enhanced image in which the extracted blood vessels in the captured image are emphasized is displayed in the enhanced image display area 122. The A plurality of processed images (a type I blood vessel extraction result and a type II blood vessel extraction result to be described later) may be displayed side by side.

  FIG. 3 shows the detailed procedure of the type setting process (step S13) according to the color resolution of the image. Here, when the format stored in the photographing apparatus 110 with the dermoscope is the JPEG format (step S131 “YES”) and the sampling factor is 4: 4: 4 (step S132 “YES”), Type II blood vessel extraction is set (step S133), and otherwise (step S131 “NO”, S132 “NO”), type I blood vessel extraction E processing is set (step S134).

  That is, when the quality of color information (color quality) is low, the type I blood vessel extraction E process is selected. When the quality of color information (color quality) is high, the type II blood vessel extraction E process is selected. Here, the type is determined based on the sampling factor in the JPEG thin line format. However, for example, even if the color resolution and the luminance resolution are the same, the type can be selected by using an uncompressed image such as a RAW image or a JPEG compressed image. It may be determined. In this case, the RAW image has high color quality, and the JPEG image has low color quality. Note that the color quality may be low when a single-plate image sensor is used, and the color quality may be high when a three-plate image sensor is used. Alternatively, it may be set manually by the user judging the color quality.

  In addition, the image conversion by the type I blood vessel extraction E process described later has an advantage that it is possible to cleanly extract even a blood vessel that can be recognized slightly on a captured image, and there is a disadvantage that it is easy to pick up noise such as bubbles and hairs. . In addition, image conversion by type II blood vessel extraction E processing is difficult to pick up noise such as bubbles and hairs, and the color quality strongly affects the quality of blood vessel extraction. Therefore, if the color quality is good, blood vessel extraction can be performed more accurately. There are advantages. Conversely, if the color quality is insufficient, there is a disadvantage that blood vessels that are slightly visible in the captured image cannot be extracted. This embodiment is intended to take advantage of these advantages according to the color quality classification of the captured image.

  A type I blood vessel extraction method will be described with reference to FIG. According to FIG. 4, in the processing unit 101, first, the separating unit 101a converts the captured image in the RGB color space into the Lab color space (more precisely, the CIE 1976 L * a * b color space) (step S151). ). Details of the Lab color space can be found on the Internet URL (http://ja.wikipedia.org/wiki/Lab%E8%89%B2%E7%A9%BA%E9%96%93) <September 1, 2014 >.

  Next, in the processing unit 101, the extraction unit 101b extracts a part selected as a diagnosis target. Specifically, the first extraction unit 101b-1 extracts the candidate for selected site (blood vessel candidate) from the luminance components separated in the Lab color space. Therefore, the first extraction unit 101b-1 uses the L image corresponding to the luminance in the Lab color space that has been color space converted by the separation unit 101a, and uses the morphological process A (first morphological process) to obtain the blood vessel candidate image BH. Is obtained (step S152). Here, the morphological processing is to apply a structuring element to an input image and generate a blood vessel candidate image BH as an output image of the same size, and each value of the output image is a corresponding pixel in the input image. Based on comparison with neighboring pixels.

  The most basic morphological processes are expansion and contraction. Dilation adds pixels to the boundary of the object in the input image, and shrinkage removes the boundary pixels. The number of pixels added to or deleted from the object varies depending on the size and shape of the structuring element used for image processing.

  Here, a method of executing the morphological process A and extracting a region (blood vessel candidate) selected as a diagnosis target from the luminance component will be described. The detailed procedure of the bottom hat process is shown in FIG.

  According to FIG. 5, the first extraction unit 101b-1 performs an expansion process on the L image to obtain a processed luminance image L1 (step S152a). Details of the expansion process can be found on the Internet URL (http://www.mathworks.co.jp/jp/help/images/morphology-fundamentals-dilation-and-erosion.html) <browse September 1, 2014> Have been described.

  Next, the first extraction unit 101b-1 performs a contraction process on the brightness image L1 after the expansion process, and obtains a brightness image L2 after the contraction process (step S152b). Subsequently, the first extraction unit 101b-1 performs smoothing filter processing for smoothing the luminance on the luminance image L2 after the contraction processing, and obtains a smoothed luminance image L3 (step S152c). Here, smoothing is performed by a Gaussian filter.

Smoothing by the Gaussian filter is expressed by the following arithmetic expression.
f (x, y) = (1 / (2πσ ^ 2)) exp (− (x ^ 2 + y ^ 2) / (2σ ^ 2))

  In the Gaussian filter, weighting by a Gaussian distribution is used as the predetermined rate. The smoothness can be controlled by the magnitude of σ in the arithmetic expression described above, and this is realized by setting a predetermined value. The smoothing filter is not limited to a Gaussian filter, and a median filter, an average filter, or the like may be used. The blood vessel candidate image BH after the bottom hat processing is obtained by subtracting the L image from the luminance image L3 after the smoothing processing (BH = L2-L) (step S152d).

  Here, the expansion process will be supplemented. For example, consider a structured element with a radius of 5 dots. The expansion processing is to perform processing for all the pixels with the maximum value within the range of the structuring element of the target pixel as the value of the target image. That is, the output target pixel value is the maximum value of all the pixels in the vicinity of the input pixel. On the other hand, in the reduction process, the minimum value within the range of the structuring element of the target pixel is set as the value of the target pixel. That is, the target pixel value is the minimum value of all the pixels in the vicinity of the input pixel. The structuring element is a circle, but may be a rectangle, for example. However, the smoothness of the smoothing filter can be reduced by using a circle.

  Returning to FIG. Next, in the processing unit 101, the second extraction unit 101b-2 extracts the likelihood (blood vessel likelihood) of the selected part using a color space composed of a luminance component and a color information component. For this reason, the second extraction unit 101b-2 calculates the likelihood of blood vessels as the likelihood A (step S153). An example of how to obtain the likelihood A is shown in FIG.

  According to FIG. 6, the processing unit 101 determines that the second extraction unit 101 b-2 has the a-axis value that is a color information component corresponding to the red hue direction of the Lab color space, and the blue hue direction. The b-axis value that is the corresponding color information component is used for extraction (step S153a). The second extraction unit 101b-2 generates LH1 by performing the following calculation from the values of the a-axis and b-axis of the Lab color space (step S153b).

ad = (a−ca) * cos (r) + b * sin (r) + ca
bd = − (a−ca) * sin (r) + b * cos (r)
LH1 = exp (-((ad * ad) / sa / sa + (bd * bd)
/ Sb / sb))

  Here, ad and bd are obtained by rotating the ab plane by r radians counterclockwise around (ca, 0). Also, the value of r is set to about 0.3 to 0.8 radians. ca is set between 0 and 50. sa and sb are the reciprocal of the sensitivity in the a-axis direction and the reciprocal of the sensitivity in the b-axis direction, respectively. Here, it is set as sa> sb.

  Next, the second extraction unit 101b-2 limits the obtained LH1 with the luminance L. If the luminance L is greater than or equal to the threshold TH1, the one set to 0 is set to LH2 (step S153c), and the luminance L is equal to or lower than the threshold TH2 is set to LH3 (step S153d). Here, the threshold value TH1 is set between 60 and 100, and the threshold value TH2 is set between 0 and 40. LH3 obtained here is set as likelihood A indicating the likelihood of blood vessels (step S153e).

  Returning to FIG. The second extraction unit 101b-2 extracts the blood vessel likelihood as the likelihood A according to the above-described procedure (step S153), and then each of the blood vessel candidate image BH after the bottom hat process and the likelihood A indicating the blood vessel likelihood. The elements are multiplied and divided by the coefficient N (step S154). Further, the blood vessel extraction E emphasized by performing clipping processing at 1 is generated (step S155).

  According to the above-described example, the blood vessel extraction E is a multi-valued image having a value from 0 to 1, but has undergone a pottom hat process, so the boundary of the extracted blood vessel is steep. If it is desired to obtain a steeper boundary, binarization may be performed with a desired threshold value.

  As described above, the second extraction unit 101b-2 uses the plane coordinates formed by the red hue direction of the color space and the blue hue direction as the center of a specific point on the red hue direction axis. Then, the likelihood A indicating the blood vessel likeness of the selected part is calculated by rotating a predetermined angle counterclockwise and limiting the luminance component in a specific value range. Then, the selected likelihood is emphasized by multiplying the calculated likelihood A by the luminance image obtained by performing the bottom hat process on the luminance component image.

  A modification in which the likelihood of blood vessels is extracted as likelihood A will be described with reference to the flowchart of FIG. The extraction unit obtains the value of the a axis that is the red hue direction of the Lab color space (step S153x), sets S to 80 (step S153y), and sets the value of the blood vessel likelihood A in the range of 0 to 80. Normalization (A ← max (min (a, S), 0) / S) is performed with a restriction, and the value is set in the range of 0 to 1 (step S153z). Here, the limit is given by a value from 0 to 80, but this value is an example and is not limited to this value.

  Next, a blood vessel extraction method according to type II, that is, a method of directly extracting blood vessels from color information will be described with reference to the flowcharts of FIGS. In the following description, a blood vessel likelihood image is generated from color information, and blood vessels are extracted by an improved top hat process (hereinafter referred to as morphological process B). The blood vessel likelihood image has a higher image value when the likelihood is higher.

  In the morphological process A shown in FIG. 4, the reduction process is performed after the expansion process is performed on the source image. A process in which expansion and contraction are repeated the same number of times is called a closing process. That is, the diagnosis support apparatus 100 according to the present embodiment performs smoothing filter processing on the image subjected to the closing processing, and subtracts (black hat processing) from the source image. Here, the source image is the luminance image L, and the value of the image in the blood vessel is relatively small. As described above, the morphological process A shown in FIG. 4 is used when extracting a shape having a small value in the image.

  Hereinafter, the blood vessel extraction E process by the morphological process B will be described. As illustrated in FIG. 8, in the processing unit 101, first, the separation unit 101a performs Lab color space conversion on a captured image in the RGB color space (step S161). Next, in the processing unit 101, the second extraction unit 101 b-2 extracts the likelihood (blood vessel likelihood) of the selected part using the color information component separated in the Lab color space. For this reason, the second extraction unit 101b-2 calculates the likelihood of blood vessels as the likelihood A (step S162). The method for obtaining the blood vessel likelihood image A is as described with reference to FIGS. In addition, you may make it remove about the blood vessel likelihood image A obtained by the color information component which has a luminance information component larger or smaller than a prescribed value. This is because if you look at the brightness of blood vessels taken under the proper shooting conditions, you will have very high brightness values or rarely very low brightness values, but the results when the conditions are not met. As described above, the upper limit is set between 60 and 100, and the lower limit is set between 0 and 40. In particular, the upper limit is preferably 80 and the lower limit is preferably set to 20.

  Subsequently, the second extraction unit 101b-2 acquires a blood vessel extraction E image from the blood vessel likelihood image A indicating the likelihood of blood vessels (step S163). The procedure for acquiring the blood vessel extraction E image from the blood vessel likelihood image A is shown in FIG.

  According to FIG. 9, first, the second extracting unit 101b-2 performs a contraction process on the blood vessel likelihood image A with an appropriate structuring element to obtain a blood vessel likelihood image A1 after the contraction process (step S163a). ). Next, the blood vessel likelihood image A1 after the contraction process is expanded to obtain a blood vessel likelihood image A2 after the expansion process (step S163b). The second extraction unit 101b-2 further applies a smoothing filter process (Gaussian filter) to the blood vessel likelihood image A2 after the expansion process, and obtains a blood vessel likelihood image A3 after the smoothing process (step S163c). Finally, a blood vessel extraction image E is obtained by subtracting the smoothed blood vessel likelihood image A3 from the blood vessel likelihood image A (step S163d).

  As described above, an image obtained by performing a contraction process on the source image (blood vessel likelihood image A) and subsequently performing an expansion process is called an opening process. The second extraction unit 101b-2 performs smoothing filter processing on the image that has been subjected to the opening processing, and subtracts the opening image from the source image (top hat processing) to extract the blood vessel shape in the source image. Here, since the source image is a blood vessel likelihood image, the value of the image is large where it seems to be a blood vessel.

  Returning to FIG. The second extraction unit 101b-2 obtains the blood vessel extraction E image from the blood vessel likelihood image A, then multiplies the blood vessel extraction E image by an appropriate coefficient K (step S164), and performs clipping processing at 1. The enhanced blood vessel extraction E is generated (step S165).

(Effect of embodiment)
As described above, according to the diagnosis support apparatus 100 according to the present embodiment, the processing unit 101 determines whether the extraction unit 101b classifies the captured image based on the color quality, and the region candidate when the color quality of the captured image is low. The first extraction means 101b-1 for extracting the image by the luminance component, and the second extraction means 101b-2 for extracting the part-likeness by the color information component when the color quality of the photographed image is high. By selectively applying the first extracting unit 101b-1 and the second extracting unit 101b-2, it is possible to extract a region candidate by an optimal method based on the color quality of the captured image, and this is displayed on the display device 120. By displaying on the screen, it is possible to support a doctor's diagnosis.

  In other words, it is possible to provide an interface for selecting a blood vessel extraction algorithm according to the color quality of a captured image, thereby facilitating a doctor's diagnosis and improving the diagnostic accuracy. Specifically, when the color quality of the captured image is low, a blood vessel candidate is extracted from the luminance information, and the blood vessel likelihood is acquired from the color information, and blood vessels are extracted from both. On the other hand, when the color quality of the captured image is high, the blood vessel likelihood is acquired from the color information, and a blood vessel is extracted from the blood vessel likelihood or the luminance component larger than the specified value is added to the blood vessel likelihood obtained from the color information. A blood vessel is extracted from a blood vessel likelihood image from which a component having a luminance less than a specified value is removed. Therefore, the blood vessel can be emphasized by an optimum method according to the quality of the captured image, and a doctor's diagnosis can be supported.

  The determination of color quality uses, for example, a sampling factor in the JPEG storage format, the type of imaging device to be used (single-plate type or three-plate type), or the presence or absence of a compressed image (irreversible compression or reversible compression). The user can also specify manually.

  As mentioned above, although this invention was demonstrated using each embodiment, it cannot be overemphasized that the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above-described embodiments and examples. Further, it is apparent from the description of the scope of claims that embodiments with such changes or improvements can be included in the technical scope of the present invention.

  The invention described in the scope of claims attached to the application of this application will be added below. The item numbers of the claims described in the appendix are as set forth in the claims attached to the application of this application.

[Appendix]
[Claim 1]
A diagnosis support apparatus for diagnosing a lesion using a photographed image of an affected area,
An image storage unit for storing the captured image;
A processing unit that processes the captured image stored in the image storage unit,
The processing unit is
Separating means for separating the captured image into a luminance component and a color information component;
An extraction means for extracting a site to be diagnosed,
The extraction means comprises:
A classifying means for classifying the captured image based on color quality;
Including at least one of first extraction means for extracting a part candidate by the luminance component, and second extraction means for extracting the likelihood of the part by a color space composed of the luminance component and the color information component,
A diagnosis support apparatus, wherein the first extraction means and the second extraction means are selectively applied according to the classification of the classification means.
[Claim 2]
The classification means is a first classification when the photographed image is an uncompressed image or an image from which color information is not thinned out, and a second classification when the photographed image is thinned out from a compressed image or color information. The diagnosis support apparatus according to claim 1.
[Claim 3]
2. The diagnosis support apparatus according to claim 1, wherein the classification unit sets the first classification when the color quality of the captured image is high, and sets the second classification when the color quality of the captured image is low.
[Claim 4]
In the first section,
The first extraction means extracts the part candidate by first morphological processing using the luminance component, and the second extraction means extracts the part likelihood using the color space,
The diagnosis support apparatus according to claim 2 or 3, wherein the extraction unit integrates the extracted part candidate and the part-likeness to generate a part extraction image.
[Claim 5]
The first morphological process includes a closing process in which an expansion process and a contraction process are repeated in this order for the extracted luminance component, the smoothing filter process for the luminance component subjected to the closing process, and the smoothing filter process. The diagnosis support apparatus according to claim 4, further comprising a subtraction process for subtracting the luminance component of the captured image from the applied luminance component.
[Claim 6]
In the second section,
The second extraction means extracts the part-likeness using the color information component,
The diagnosis support apparatus according to claim 2 or 3, wherein the extraction unit generates a part extraction image by a second morphological process using the extracted part-likeness.
[Claim 7]
In the second morphological process, an opening process for repeating the contraction process and an expansion process in this order on the extracted part-likeness, the smoothing filter process on the part-likeness on which the opening process has been performed, and the smoothing filter process are performed. The diagnosis support apparatus according to claim 6, further comprising a process of subtracting the extracted part likelihood from the extracted part likelihood.
[Claim 8]
The second morphological processing generates a part extraction image from the part-likeness obtained by removing those having a luminance component larger than a prescribed value or those having a luminance component less than a prescribed value with respect to the extracted part-likeness. The diagnosis support apparatus according to claim 6.
[Claim 9]
An image processing method in a diagnosis support apparatus for diagnosing a lesion using a photographed image of an affected area,
Separating the stored captured image into a luminance component and a color information component (A);
A step (B) of classifying the captured image based on color quality;
(C) extracting a site to be diagnosed by the luminance component or the color information component,
When the color quality is classified as high in the classification step (B),
The extraction step (C) executes a first extraction step of extracting a part candidate by the color information component,
When the color quality is classified as low in the classification step (B),
The image processing method according to claim 1, wherein the extraction step (C) executes a second extraction step of extracting a part-likeness by a color space composed of the luminance component and the color information component.
[Claim 10]
An image processing program in a diagnosis support apparatus for diagnosing a lesion using a photographed image of an affected area,
On the computer,
A separation function for separating the stored captured image into a luminance component and a color information component;
A sorting function for sorting the captured image based on color quality;
An extraction function for extracting a region to be diagnosed by the luminance component or the color information component; and
When the color function is classified as high in the classification function,
The extraction function performs a first extraction function of extracting a region candidate by the color information component;
When the color quality is classified as low in the classification function,
The program according to claim 1, wherein the extraction function executes a second extraction function for extracting a part-likeness by a color space composed of the luminance component and the color information component.

DESCRIPTION OF SYMBOLS 100 ... Diagnosis assistance apparatus, 101 ... Processing part, 101a ... Separation means, 101b ... Extraction means (101b-1 ... 1st extraction means, 101b-2 ... 2nd extraction means), 110 ... Imaging apparatus with a dermoscope, 120 ... Display device, 121 ... Photographed image display area, 122 ... Enhanced image display area, 130 ... Input device

Claims (10)

A diagnosis support apparatus for diagnosing a lesion using a photographed image of an affected area,
An image storage unit for storing the captured image;
A processing unit that processes the captured image stored in the image storage unit,
The processing unit is
Separating means for separating the captured image into a luminance component and a color information component;
An extraction means for extracting a site to be diagnosed,
The extraction means comprises:
A classifying means for classifying the captured image based on color quality;
Including at least one of first extraction means for extracting a part candidate by the luminance component, and second extraction means for extracting the likelihood of the part by a color space composed of the luminance component and the color information component,
A diagnosis support apparatus, wherein the first extraction means and the second extraction means are selectively applied according to the classification of the classification means.   The classification means is a first classification when the photographed image is an uncompressed image or an image from which color information is not thinned out, and a second classification when the photographed image is thinned out from a compressed image or color information. The diagnosis support apparatus according to claim 1.   2. The diagnosis support apparatus according to claim 1, wherein the classification unit sets the first classification when the color quality of the captured image is high, and sets the second classification when the color quality of the captured image is low. In the first section,
The first extraction means extracts the part candidate by first morphological processing using the luminance component, and the second extraction means extracts the part likelihood using the color space,
The diagnosis support apparatus according to claim 2 or 3, wherein the extraction unit integrates the extracted part candidate and the part-likeness to generate a part extraction image.   The first morphological process includes a closing process in which an expansion process and a contraction process are repeated in this order for the extracted luminance component, the smoothing filter process for the luminance component subjected to the closing process, and the smoothing filter process. The diagnosis support apparatus according to claim 4, further comprising a subtraction process for subtracting the luminance component of the captured image from the applied luminance component. In the second section,
The second extraction means extracts the part-likeness using the color information component,
The diagnosis support apparatus according to claim 2 or 3, wherein the extraction unit generates a part extraction image by a second morphological process using the extracted part-likeness.   In the second morphological process, an opening process that repeats a contraction process and an expansion process in this order on the extracted part-likeness, the smoothing filter process on the part-likeness that has been subjected to the opening process, and the smoothing filter process are performed. The diagnosis support apparatus according to claim 6, further comprising a process of subtracting the extracted part likelihood from the extracted part likelihood.   The second morphological processing generates a part extraction image from the part-likeness obtained by removing those having a luminance component larger than a prescribed value or those having a luminance component less than a prescribed value with respect to the extracted part-likeness. The diagnosis support apparatus according to claim 6. An image processing method in a diagnosis support apparatus for diagnosing a lesion using a photographed image of an affected area,
Separating the stored captured image into a luminance component and a color information component (A);
A step (B) of classifying the captured image based on color quality;
(C) extracting a site to be diagnosed by the luminance component or the color information component,
When the color quality is classified as high in the classification step (B),
The extraction step (C) executes a first extraction step of extracting a part candidate by the color information component,
When the color quality is classified as low in the classification step (B),
The image processing method according to claim 1, wherein the extraction step (C) executes a second extraction step of extracting a part-likeness by a color space composed of the luminance component and the color information component. An image processing program in a diagnosis support apparatus for diagnosing a lesion using a photographed image of an affected area,
On the computer,
A separation function for separating the stored captured image into a luminance component and a color information component;
A sorting function for sorting the captured image based on color quality;
An extraction function for extracting a region to be diagnosed by the luminance component or the color information component; and
When the color function is classified as high in the classification function,
The extraction function executes a first extraction function of extracting a part candidate by the color information component;
When the color quality is classified as low in the classification function,
The program that causes the extraction function to execute a second extraction function that extracts a part-likeness by a color space constituted by the luminance component and the color information component.

Images